Digging around some more, wanted to share what I found:

Using this link - http://support.microsoft.com/kb/923247

I did a DBCC PAGE on the affected pages and went to the specific slot that reported an issue. Then I was able to find the pobs_id of that row and could do a SELECT against the table and see the 'bad data':

DBCC PAGE ( pmdb_cn , 1 , 110117 , 3 )

Slot 38 Column 3 Offset 0x3c Length 255 Length (physical) 255

pobs_name = INVALID COLUMN VALUE

select * from POBS where pobs_id=6932

????????????????????????????????????????????????????????????????????????????????????????????????????????????????????????????????

So it looks like i have 4 rows where the data is unicode (Chinese characters) and cannot fit into this column.

I'm thinking that the datatype of the column should be NVARCHAR not VARCHAR in order to accept these values and be compatible with unicode characters. The db collation is Chinese so i *think* this column should be set to nvarchar..thoughts?